intronic variants in the nfkb1gene may influence...
TRANSCRIPT
Intronic Variants in the NFKB1 Gene May InfluenceHearing Forecast in Patients with UnilateralSensorineural Hearing Loss in Meniere’s DiseaseSonia Cabrera1, Elena Sanchez2, Teresa Requena1, Manuel Martinez-Bueno3, Jesus Benitez4,
Nicolas Perez5, Gabriel Trinidad6, Andres Soto-Varela7, Sofıa Santos-Perez7, Eduardo Martin-Sanz8,
Jesus Fraile9, Paz Perez10, Marta E. Alarcon-Riquelme3, Angel Batuecas11, Juan M. Espinosa-Sanchez1,12,
Ismael Aran13, Jose A. Lopez-Escamez1,14*
1 Otology & Neurotology Group CTS495, Department of Genomic Medicine- Centro de Genomica e Investigacion Oncologica – Pfizer/Universidad de Granada/Junta de
Andalucıa (GENYO), Granada, Spain, 2 Department of Neurology, Icahn School of Medicine at Mount Sinai, New York City, New York, United States of America, 3 Group of
Genetics of Complex Diseases, Department of Genomic Medicine - Centro de Genomica e Investigacion Oncologica – Pfizer/Universidad de Granada/Junta de Andalucıa
(GENYO), Granada, Spain, 4 Department of Otolaryngology, Hospital Universitario de Gran Canaria Dr Negrin, Las Palmas, Spain, 5 Department of Otolaryngology, Clinica
Universidad de Navarra, Pamplona, Spain, 6 Division of Otoneurology, Department of Otorhinolaryngology, Complejo Hospitalario Badajoz, Badajoz, Spain, 7 Division of
Otoneurology, Department of Otorhinolaryngology, Complexo Hospitalario Universitario, Santiago de Compostela, Spain, 8 Department of Otolaryngology, Hospital
Universitario de Getafe, Getafe, Spain, 9 Department of Otolaryngology, Hospital Miguel Servet, Zaragoza, Spain, 10 Department of Otorhinolaryngology, Hospital
Cabuenes, Gijon, Spain, 11 Department of Otolaryngology, Hospital Universitario Salamanca, Salamanca, Spain, 12 Department of Otorhinolaryngology, Hospital San
Agustin, Linares, Jaen, Spain, 13 Department of Otolaryngology, Complexo Hospitalario de Pontevedra, Pontevedra, Spain, 14 Department of Otolaryngology, Hospital de
Poniente, El Ejido, Almerıa, Spain
Abstract
Meniere’s disease is an episodic vestibular syndrome associated with sensorineural hearing loss (SNHL) and tinnitus. Patientswith MD have an elevated prevalence of several autoimmune diseases (rheumatoid arthritis, systemic lupus erythematosus,ankylosing spondylitis and psoriasis), which suggests a shared autoimmune background. Functional variants of severalgenes involved in the NF-kB pathway, such as REL, TNFAIP3, NFKB1 and TNIP1, have been associated with two or moreimmune-mediated diseases and allelic variations in the TLR10 gene may influence bilateral affectation and clinical course inMD. We have genotyped 716 cases of MD and 1628 controls by using the ImmunoChip, a high-density genotyping arraycontaining 186 autoimmune loci, to explore the association of immune system related-loci with sporadic MD. Although nosingle nucleotide polymorphism (SNP) reached a genome-wide significant association (p,1028), we selected allelic variantsin the NF-kB pathway for further analyses to evaluate the impact of these SNPs in the clinical outcome of MD in our cohort.None of the selected SNPs increased susceptibility for MD in patients with uni or bilateral SNHL. However, two potentialregulatory variants in the NFKB1 gene (rs3774937 and rs4648011) were associated with a faster hearing loss progression inpatients with unilateral SNHL. So, individuals with unilateral MD carrying the C allele in rs3774937 or G allele in rs4648011had a shorter mean time to reach hearing stage 3 (.40 dB HL) (log-rank test, corrected p values were p = 0.009 forrs3774937 and p = 0.003 for rs4648011, respectively). No variants influenced hearing in bilateral MD. Our data support thatthe allelic variants rs3774937 and rs4648011 can modify hearing outcome in patients with MD and unilateral SNHL.
Citation: Cabrera S, Sanchez E, Requena T, Martinez-Bueno M, Benitez J, et al. (2014) Intronic Variants in the NFKB1 Gene May Influence Hearing Forecast inPatients with Unilateral Sensorineural Hearing Loss in Meniere’s Disease. PLoS ONE 9(11): e112171. doi:10.1371/journal.pone.0112171
Editor: David R. Booth, University of Sydney, Australia
Received June 6, 2014; Accepted October 13, 2014; Published November 14, 2014
Copyright: � 2014 Cabrera et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permitsunrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The authors confirm that all data underlying the findings are fully available without restriction. All relevant data are within the paper and itsSupporting Information files.
Funding: This project was funded by Research Grants PI09/00920 and PI13/1242 from Instituto de Salud Carlos III by FEDER Funds from the EU. The authors alsoacknowledge the COST Action BM1306 TINNET which supports part of their networking activities (http://tinnet.tinnitusresearch.net/). The funders had no role instudy design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* Email: [email protected]
Introduction
Meniere’s disease (MD) is a chronic disorder affecting the inner
ear characterized by fluctuating sensorineural hearing loss
(SNHL), episodes of vertigo, tinnitus, and aural fullness and it
can affect both ears in 10–40% of cases [1]. The etiology and
pathogenesis remain unknown, although one-third of MD cases
may have an aberrant response of the adaptive or innate immune
system, the immunological mechanisms involved have not been
investigated [2].
Several mechanisms may explain the development of immune-
mediated inner ear disease (IED): a) cross-reactions with a cross-
reactive epitope (antibodies cause accidental inner ear damage
because the ear shares epitopes with a potentially harmful
substance, virus or bacteria) as suspected for some inflammatory
diseases [3], b) damage to the inner ear caused by pro-
PLOS ONE | www.plosone.org 1 November 2014 | Volume 9 | Issue 11 | e112171
inflammatory cytokines such as IL-1B [4], [5] or TNF [6] as in
some autoimmune diseases, or c) inappropriate immune response
or intolerance to harmless unrecognized substances combined with
genetic factors that modify the immune response as in allergies [2].
IED and MD may have an overlapping phenotype and
autoimmune mechanisms could be associated with the pathophys-
iology of MD. Some evidence support this hypothesis including the
response to steroids therapy, the finding of elevated levels of
autoantibodies or circulating immune complexes (CIC) in the
serum of some patients with MD against inner ear antigens and
the association of allelic variants of MHC class I polypeptide-
related sequence A (MICA) and Toll-like receptor 10 (TLR10,
rs11096955) gene with hearing loss progression in patients with
MD [7], [8].
These findings together with the elevated prevalence of several
autoimmune diseases such as rheumatoid arthritis (RA), systemic
lupus erythematosus (SLE), ankylosing spondylitis (AS), and
psoriasis in patients with MD, suggest an autoimmune component
[9], [10].
Nuclear factor kB (NF-kB) is a crucial pleiotropic transcription
factor (TF) which regulates inflammation and the innate and
adaptive immune response [11]. Five members of this transcrip-
tion factor family have been identified: RELA (p65), RELB, REL
(c-Rel), NF-kB1 (p105) and NF-kB2 (p100). Of note, p105 and
p100 are pro-forms proteolytically processed to p50 and p52 [12].
All members of the NF-kB family harbor an N-terminal Rel
homology domain (RHD), which interacts with DNA elements
and mediates homo- and heterodimerization. The complex p65–
p50 is the most abundant form of heterodimer of the NF-kB family
and it keeps in an inactive state in the cytoplasm bound to proteins
of the IkB family, which are inhibitors of NF-kB [13].
In its canonical pathway, excitatory signaling can be mediated
through Toll-like receptors (TLRs), Interleukin-1 receptor (IL-1R),
tumor necrosis factor receptor (TNFR) and antigen receptors.
Several genes involved in the regulation of NF-kB pathway have
been associated with autoimmune disorders. So, ubiquitin-
conjugating enzyme E2L3 (UBE2L3), which ubiquitylates p105
for its degradation, tumor necrosis factor alpha-induced protein 3
(TNFAIP3, also known as A20), an ubiquitin-editing enzyme with
determines NF-kB activity or TNFAIP3-interacting protein
(TNIP1) which inhibits TNF-induced NF-kB -dependent gene
expression have genetic variants associated with several autoim-
mune diseases [14].
The genotyping of large cohorts of patients with several
autoimmune diseases in genome-wide association studies has
shown that most of these diseases share multiple susceptibility loci
(www.immunobase.org) [14]. Among them, several genes in the
NF-kB pathway are associated with two or more immune-
mediated diseases, such as inflammatory bowel disease (REL,
TNFAIP3 and NFKB1), psoriasis (REL, TNFAIP3, NFKB1 and
TNIP1), coeliac disease (REL and TNFAIP3), rheumatoid
arthritis (REL and TNFAIP3), type 1 diabetes (T1D) (TNFAIP3),
systemic lupus erythematous (SLE) (TNFAIP3 and TNIP1),
multiple sclerosis and primary biliary cirrhosis (NFKB1) (Table 1).
We have used the ImmunoChip, a high-density genotyping
array which includes 186 loci previously associated with 12
autoimmune diseases, to explore the association of these loci with
MD and to evaluate the role of functional variants of genes
involved in the NF-kB pathway with MD and their potential effect
on the hearing outcome of the disease [14].
Materials and Methods
Study samplesThis study was approved by the ethics committees of all the
recruiting centers and all participating individuals gave written
informed consent. This study was approved by the ethics
committee for clinical research from Almerıa (Comite Etico de
Investigacion provincial de Almerıa), Jaen (Comite Etico de
Investigacion provincial de Jaen), Galicia (Comite Autonomico de
Etica de la Investigacion de Galicia), Asturias (Comite Etico de
Investigacion Clınica Autonomico de Asturias), Las Palmas (CEIC
del Hospital de Gran Canaria Dr. Negrin), Navarra (Comite Etico
de Investigacion Clınica Autonomico de Navarra), Extremadura
(Comite Etico Autonomico de Extremadura), Madrid-Getafe
(CEIC Area 10 - Hospital Universitario de Getafe), Aragon
(Comite Etico de Investigacion Clınica de Aragon) and Salamanca
(Comite Etico de Investigacion Clınica Area de Salud de
Salamanca). All the procedures described were performed in
accordance with the highest ethical standards on human
experimentation, the Helsinki Declaration of 1975.
We recruited a total of 716 patients who were diagnosed with
definite MD according to the diagnostic scale for MD of the
American Academy of Otolaryngology Head and Neck Surgery
(AAO-HNS) [15] and 1628 control volunteers in a case–control
study.
The diagnosis of MD was established according to the clinical
guidelines defined by the Committee on Hearing and Equilibrium
of the AAO-NHS in 1995 [15]. A complete neuro-otological
evaluation including otoscopy, a pure-tone audiometry, nystagmus
examination and caloric testing was carried out in all cases.
Moreover, the protocol of diagnosis included a brain MRI to
exclude other possible causes of neurological symptoms. Patients
were followed with serial audiograms at each visit to monitor
hearing loss from the initial diagnosis. The following clinical
variables were studied in our series: gender, age, hearing stage,
duration of the disease, bilateral SNHL, age of onset, type of
headache, history of autoimmune disease, smoking, Tumarkin
crisis and the functional scale of the AAO-HNS. Hearing staging
was calculated by the audiogram obtained the day of inclusion for
each patient with definite MD and was defined as the mean of
four-tone average of 0.5, 1, 2 and 3 kHz according to the AAO-
HNS criteria: stage 1, #25 dB; stage 2, 26–40 dB; stage 3, 41–
70 dB; stage 4, .70 dB.
DNA extraction and genotypingBlood samples from each subject were collected and genomic
DNA was isolated from peripheral blood leukocytes using the
QIAamp DNA Mini Kit (Quiagen), according to the manufac-
turer’s instructions. All genomic DNA was re-suspended in
nuclease free water for the following study.
The concentration of genomic DNA was measured using the
Qubit dsDNA BR Assay Kit (Invitrogen) and concentrations were
standardized to 50 ng/mL for genotyping. All samples were
genotyped using the ImmunoChip, a custom Illumina Infinium
high density genotyping array containing 196524 markers across
186 known autoimmunity risk loci [14],in the iScan genotyping
platform (Illumina Inc., San Diego, CA).
Quality controlsSamples were clustered together by using the Illumina Genome
Studio algorithm. Clusters were manually inspected and verified,
removal of single nucleotide polymorphisms (SNPs) with poor
clustering quality metrics (call frequency ,0.98, cluster separation
,0.4) was performed, and all SNPs with GenCall scores less than
NFKB1 Gene and Hearing in Meniere’s Disease
PLOS ONE | www.plosone.org 2 November 2014 | Volume 9 | Issue 11 | e112171
0.15 were excluded. Quality controls (QC) were applied for all
individuals and genotyped SNPs by using PLINK software
(version 1.07) [16].
Samples with a genotype success rate of ,90% were excluded
from the analysis. The remaining samples were then evaluated for
duplicates or related individuals and one individual from each pair
was removed if the proportion of alleles share identical by descent
(IBD) .0.5. Samples with increased heterozygosity rate (,0.18
and . 0.45) and missing data between cases and controls P value
,1025, were then removed from the analysis. Finally, genetic
outliers determined by principal component analysis (PCA) were
removed from the analysis (. 3 standard deviation around the
mean). All familial cases were also excluded.
All the SNPs that did not meet the following criteria were
excluded from further analysis: genotype success rate ,90%,
minor allele frequency (MAF) ,5%, Hardy-Weinberg equilibrium
,1024 in controls and missing-genotype rate ,0.5%. All markers
Table 1. Single nucleotide polymorphisms in the NF-kB pathway with reported associations.
Chr Positiona rsID GENE (variant type) Phenotype Association P-Value Reference
6 138199417 rs610604 TNFAIP3 (intron) Psoriasis 5.5361025 [39]
5 150440097 rs2233287 TNIP1(intron) Asthma, systemic sclerosis 0.039, 6.1761024 [40,41]
5 150438988 rs1422673 TNIP1(intron) Asthma 0.011 [40]
2 61136129 rs13031237 REL (intron) Rheumatoid Arthritis 7.2961023 [42]
22 21939675 rs5754217 UBE2L3 (intron) SLE 0.012 [43]
22 21917190 rs131654 UBE2L3 (intron) SLE 1.1261027 [44]
22 21932068 rs181362 UBE2L3 (intron) HDL cholesterol 3.7261024 [45]
4 103434253 rs3774937 NFkB1 (intron) Primary biliary cirrhosis, Body Weight 1.5610210 0.041 [46, dbGaPb]
4 103475444 rs4648011 NFkB1 (intron) Body Weight 0.040 [dbGaPb]
aNCBI human genome build 37 coordinates.bhttp://www.ncbi.nlm.nih.gov/gap [38].doi:10.1371/journal.pone.0112171.t001
Table 2. Clinical features of patients with Meniere’s disease and uni or bilateral sensorineural hearing loss.
VARIABLES BILATERAL (n = 168) UNILATERAL (n = 548) P-value
Age of onset, mean ± SD 46.6612.5 46.9612.1 0.743
Gender (% women) 60.4 56.6 0.404
Time course (years), mean ± SD 11.268.7 7.966.7 1.561025
Affected ear (%) Left (50.6) Right (49.4)
Hearing loss at diagnosis, mean ± SD 53.9616.6 48.9617.3 0.003
Migraine, n (%) 25 (14.8) 56 (10.2) 0.719
History autoimmune disease, n (%) 36 (21.4) 62 (11.3) 0.003
Smoking, n (%) 40 (23.8) 134 (29.9) 0.882
Hearing stage, n (%)
1 7 (4.2) 58 (12.6) 8.061026
2 28 (16.6) 116 (21.2)
3 78 (46.4) 260 (47.4)
4 48 (28.5) 69 (12.5)
Hearing stage, mean ± SD 3.0160.8 2.6860.9 3.061026
Turmakin crisis, n (%) 44 (26.1) 77 (14.1) 0.001
Functional scale, n (%)
1 28 (16.6) 98 (17.9) 0.964
2 46 (27.3) 158 (28.8)
3 36 (21.4) 104 (18.9)
4 25 (14.8) 79 (14.4)
5 16 (9.5) 46 (8.3)
6 3 (1.7) 7 (1.2)
Age of onset, time course years and hearing loss at diagnosis were compared by unpaired Student’s t test. Qualitative variables were compared byChi-squared test.doi:10.1371/journal.pone.0112171.t002
NFKB1 Gene and Hearing in Meniere’s Disease
PLOS ONE | www.plosone.org 3 November 2014 | Volume 9 | Issue 11 | e112171
in chromosome X were also excluded. After QC, 96899 SNPs
remained with a MAF.0.05 for statistical analysis.
Statistical analysisAfter all QC, 689 cases (521 unilateral, 168 bilateral) and 1475
controls remained for further analyses. We have evaluated the
association between each SNPs and patients with uni or bilateral
MD. Allelic and genotype frequencies were compared between
patients and controls by using x2 test and calculating the odds
ratios (OR) and 95% confidence intervals (CIs) using PLINK
(version 1.07). P-values were adjusted by genomic control. The
power was computed as the probability of detecting an association
at the 0.05 significance level, assuming an OR = 1.5 (small effect
size). Power analysis was estimated using the Quanto v1.2.4
software (Department of Preventive Medicine University of
Southern California, CA, USA).
Moreover, among the SNPs compared above, we selected those
SNPs which have been previously associated with other immune-
mediated diseases, in the following genes: NFKB1 (rs3774937,
rs4648011), REL (rs13031237), UBE2L3 (rs5754217, rs131654,
rs181362), TNFAIP3 (rs610604) and TNIP1 (rs2233287,
rs1422673) (Table 1). To assess if any of these variants have any
effect on the clinical course of patients with MD. The median time
to reach hearing loss .40 dB (hearing stage 3 or 4) for each
genotype/allele was calculated according to the Kaplan–Meier
method and survival curves were compared using the log-rank test
(IBM SPSS Statistics 20.0). The p values were corrected, according
to the Bonferroni’s method, for the number of comparison made
for each gene. p,0.05 was considered statistically significant.
Validation of rs3774937 and rs4648011 genotypingCase samples were also genotyped for two SNPs in the NFKB1
gene (rs3774937 and rs4648011) with a TaqMan 59 allelic
discrimination assay according to manufactures’ instructions (Life
Technologies). Amplifications were performed in an ABI 7500
Fast Real-Time PCR System (LT) for continuous fluorescence
monitoring. The alleles were determined using the SDS 2.2.1
software (LT). Functional evaluation of these regulatory variants
was performed in silico by using the bioinformatics tools
HaploReg (http://www.broadinstitute.org/mammals/haploreg/
haploreg.php), seeQTL (http://www.bios.unc.edu/research/
genomic_software/seeQTL/) and RegulomeDB (http://
regulomedb.org/) to explore annotations of the noncoding
genome such as candidate regulatory SNPs, conservation across
mammals and its potential effects on regulatory motifs [17–19].
Results
Table 2 compares the basic clinical features of 716 patients with
uni and bilateral MD in our series. As we expected, patients with
bilateral SNHL had a longer duration of disease (p = 1.561025),
worse hearing loss at diagnosis (p = 0.003) and worse hearing stage
(p = 361026), a higher frequency of Tumarkin crises (p = 0.001)
and autoimmune disease comorbidities (p = 0.003). However, no
differences were observed in the age of onset or frequency of
migraine between patients with uni or bilateral SNHL.
Principal-component analysis (PCA) showed that cases and
controls had similar distributions of the top two eigenvectors in
both sets, suggesting a common genetic background for these study
subjects (Figure S1 in File S1). We found no evidence of
population stratification in the PCA.
Table 3. Effect of allelic variants in the NF-kB pathway on hearing loss progression in patients with Meniere’s disease.
Long-rank test (p-value)
GENE SNV UNILATERAL BILATERAL
TNFAIP3 (intron) rs610604 genotype 0.469 0.431
allele 0.045 0.385
TNIP1 (intron) rs2233287 genotype 0.806 0.505
allele 0.612 0.521
TNIP1 (intron) rs1422673 genotype 0.849 0.729
allele 0.303 0.365
REL (intron) rs13031237 genotype 0.276 0.053
allele 0.613 0.449
UBE2L3 (intron) rs5754217 genotype 0.779 0.276
allele 0.992 0.360
UBE2L3 (intron) rs131654 genotype 0.738 0.208
allele 0.606 0.196
UBE2L3 (intron) rs181362 genotype 0.779 0.276
allele 0.992 0.360
NFKB1 (intron) rs3774937 genotype (CC) 0.018* 0.160
allele (C) 0.009* 0.995
NFKB1 (intron) rs4648011 genotype (GG) 0.018* 0.420
allele (G) 0.003* 0.956
Mean time to reach stage 3 (.40 dB) was compared by Kaplan –Meier survival curves and long-rank test.*corrected p values after Bonferroni’s method.doi:10.1371/journal.pone.0112171.t003
NFKB1 Gene and Hearing in Meniere’s Disease
PLOS ONE | www.plosone.org 4 November 2014 | Volume 9 | Issue 11 | e112171
No single marker reached a genome-wide significant (p,1028)
when all cases and controls were compared. The top ten signals
found in patients with MD are shown in Table S1 in File S1. The
allelic frequencies of the selected variants in genes NFKB1, REL,UBE2L3, TNFAIP3 and TNIP1 in patients and controls are
shown in Tables S2, S3, S4, S5, and S6 in File S1. There was no
significant difference among patients with uni or bilateral SNHL
for any of the SNPs studied (p.0.05). We also stratified patients in
two groups according to the presence of uni or bilateral SNHL
and compared each group with controls, but none of the allelic
variants reached a genome-wide significance. Moreover, none of
the selected variants were associated with comorbidities such as
autoimmune disorder or migraine.
We also analyzed the time course of hearing loss in patients with
uni or bilateral SNHL for all the functional allelic variants
previously selected in the NF-kB pathway. Kaplan-Meier analysis
showed that functional allelic variants on REL, TNFAIP3, RELand TNIP1 genes did not influence the auditory prognosis in MD
(Table 3). However, two SNP in the NFKB1 gene (rs3774937 and
rs4648011) were associated with a faster hearing loss progression
in patients with unilateral SNHL (n = 490). So, patients carrying C
allele in rs3774937 and G allele in rs4648011, respectively,
reduced in 2 years the mean time to reach hearing stage 3 (.
40 dB HL) (log-rank test, corrected p values were p = 0.009 for
rs3774937 and p = 0.003 for rs4648011, respectively; Figure 1).
So, the median of years to reach hearing stage 3 was 8, 8 or 11
years since the onset of disease for carriers of the genotype CC,
CT or TT in rs3774937, respectively (log-rank test, corrected p
value was p = 0.018). For rs4648011, we also found that the
median time to reach stage 3 was 7, 8 or 11 year since the onset of
disease for carriers of the genotype GG, GT or TT, respectively
(log-rank test, corrected p value p = 0.018). Remarkably, these
variants in the NFKB1 gene did not influence hearing in patients
with bilateral SNHL (p.0.05, 2N = 290). The allelic frequencies
of rs3774937-C and rs4648011-G were 0.31 and 0.37, respective-
ly. Both SNPs were validated in all cases by Taqman assays and
the correlation coefficient between both methods was 98%. These
variants showed high linkage disequilibrium (r2 = 0.67, D’ = 0.95;
Figure 2). The haplotype CG has a frequency of 32% and carriers
of this haplotype reached 40 dB 30 months earlier than the rest of
the haplotypes carriers (p = 0.002, Table 4). We also compared if
carriers of the CG haplotype had a faster hearing loss progression
than patients with either rs3774937-C or rs4648011-G alleles, but
no additive effect was found.
Finally, in silico analysis of these variants predicted changes in
the interaction with the following transcription factors: DMRT1,
LUN-1, YY1 for rs3774937 and Foxc1, Zfx for rs4648011.
Discussion
MD is probably a syndrome including a heterogeneous group of
patients with an immune-mediated disease and non-immune
mediated mechanisms. The 1995 clinical definition of the AAO-
HNS does not discriminate these clinical variants [15].
Our results show that allelic variants in the NFKB1 gene
influence the hearing outcome in patients with unilateral MD.
Although these markers are not associated with an increased
susceptibility to develop MD, they probably modify the interaction
of NFKB1 with other transcription factors conditioning the
inflammatory response in the inner ear. Our study has enough
power to detect an association between rs3774937 and rs4648011
and unilateral SNHL patients (.98% for both SNPs). However,
the lack of association found in bilateral SNHL patients can be due
to the smaller sample size in this cohort, resulting in a lack of
power to detect a susceptibility association (power 70% and 66%,
respectively). These intronic markers in the NFKB1 gene are
located in chromosome 4q24 and they have been strongly
associated with primary biliary cirrhosis [45].
Genetic association studies in MD have been limited to case-
control studies based on candidate-gene approaches in small series
Figure 1. Variants in NFKB1 gene and hearing outcome in patients with MD were compared by Kaplan-Meier survival curves and thelog-rank test. A. Carriers of the C allele in rs3774937 showed a shorter time to reach hearing stage 3 (.40 dB). B. Carriers of the G allele in rs4648011also reduced in 2 years the mean time to reach hearing stage 3 (log-rank test, p = 0.009 for rs3774937 and p = 0.003 for rs4648011).doi:10.1371/journal.pone.0112171.g001
NFKB1 Gene and Hearing in Meniere’s Disease
PLOS ONE | www.plosone.org 5 November 2014 | Volume 9 | Issue 11 | e112171
with low power and replication studies have failed to confirm
previous associations [20–22]. The introduction of technology of
genotyping-based arrays after finishing the Human Genome
Project [23] and the International HapMap Project [24] have
changed the approach for gene discovery in complex diseases to
large-scale testing where every gene in the human genome is tested
for association with a disease of interest.
The Immunochip project was a collaborative Consortium
among 12 immune-mediated disease groups (autoimmune thyroid
disease, ankylosing spondylitis, celiac disease, Crohn’s disease, IgA
deficiency, multiple sclerosis, primary biliary cirrhosis, psoriasis,
rheumatoid arthritis, SLE, type 1 diabetes and ulcerative colitis)
and the Welcome Trust Case Control Consortium (WTCCC).
The result was a high-density array containing 186 distinct
susceptibility loci associated with one or more immune-mediated
diseases [14]. The ImmunoChip has revealed mainly common
variants of modest effect size with odds ratios between 1.04 and
3.99 (mean = 1.29), when excluding the MHC region on most of
the immune-mediated diseases listed above [25–28] and other
diseases such as atopic dermatitis [29] or Behcet’s disease [30].
We have performed an exploratory genotyping study by using
the ImmunoChip including 716 patients and 1628 controls (689
and 1475, respectively after QC), but the clinical heterogeneity of
MD anticipates that this sample size is not enough to define the
susceptibility loci associated with MD. Previous findings and these
results suggest that patients with MD and uni and bilateral SNHL
have different genomic background since patients with bilateral
SNHL and MD have more frequent comorbid autoimmune
diseases [9], [10].
Autoimmunity has been proposed as a mechanism in patients
with bilateral SNHL and MD [2] and although preliminary
candidate gene studies found that bilateral MD was associated
with allelic variants of genes HLA-DRB1 [31], PTPN22 [32],
these findings have not been formally replicated in the current
study including 168 patients with bilateral MD, the largest
collection so far.
There are several reasons to explain our findings. The great
clinical heterogeneity of MD advances that several thousands of
patients will be required to perform a genome-wide association
study with enough power to find susceptibility loci with p values ,
1028. Although those numbers are possible by an international
multicenter study for unilateral MD, it is a real challenge to recruit
those numbers for bilateral MD.
Different studies suggest the role of innate immune response in
the hearing outcome of autoimmune inner ear disease [4], [6] and
MD [8]. So, in the present study we have found that two
regulatory variants of NFKB1 gene also influence long-term
progression of hearing loss in unilateral MD. Bioinformatics tools
predicted that these variants changed the interaction with the
following transcription factors: DMRT1, LUN-1, YY1 for
rs3774937 and Foxc1, Zfx for rs4648011. Since the NF-kB
pathway regulates proinflammatory cytokine production and cell
Figure 2. Linkage disequilibrium plot showing the haploblocks with the rs3774937 and rs4648011 (r2 = 0.67, D’ = 0.95).doi:10.1371/journal.pone.0112171.g002
NFKB1 Gene and Hearing in Meniere’s Disease
PLOS ONE | www.plosone.org 6 November 2014 | Volume 9 | Issue 11 | e112171
survival and it mediates the duration of the inflammatory
response, these variants probably influence gene expression and
inflammation in MD. However, further studies with conditioned
cells (i.e., lymphoblasts with the homozygous variants) will be
necessary to define the effect of these variants in the gene
expression profile, since multiple interactions are possible in the
NF-kB network and the molecular mechanism remains unknown.
We have recently found that rs11096955 in TLR10 gene may
confer susceptibility to bilateral SNHL in patients with MD [8].
TLRs constitute one of primary defense mechanisms in infections
and some noninfectious diseases [33] and inadequate activation of
TLRs pathway has been reported in several autoimmune diseases
[34]. Since TLRs activates an intracellular signal via MyD88,
triggering a complex cascade (IRAK1-IRAK4-TRAF6, TAK1-
TAB1-TRAF6-UBC13, IKK complex) that leads to the induction
of a large range of proinflammatory genes via the transcription
factors NF-kB [35], we have selected functional variants in the
genes NFKB1, REL, UBE2L3, TNFAIP3 and TNIP1 to
evaluate their potential role in the outcome of MD. Overall, we
suggest that allelic variants in some genes of the innate immune
response such as TLR10 and NFKB1 may act as regulatory genes
able to modify the clinical progression of hearing in MD. By using
an NF-kB reporter mouse, it was demonstrated that the protective
action of NF-kB was exerted in connective tissue cells within the
spiral ligament. So, in the spiral ligament, type II fibrocytes are
activated following systemic inflammatory stress and immune-
mediated SNHL in humans may result in part from susceptibility
of type II fibrocytes [36]. An abnormal function of type II
fibrocytes would likely have a significant impact upon hearing
thresholds, since these cells have a critical role in K+ ion uptake
from perilymph. Moreover, the fact that steroids are potent
blockers of NF-kB pathway may explain the observed response to
systemic steroids in patients with sudden SNHL or MD.
Our study design has some limitations: the sample size is not
enough to detect differences for SNPs with MAF,0.05; however is
has been recently demonstrated that the effect of rare variants in
autoimmune loci is negligible [37]. We have only compared SNPs
included in the ImmunoChip and it is necessary to scan the entire
genome in complex diseases. Moreover, the great heterogeneity of
the disorder may raise concerns about the cost-effectiveness of this
approach. An alternative approach is to select multicase families
with MD and combine whole-exome sequencing with segregation
analysis to define novel or rare variants in candidate genes.
Genomic data fusion including phenotype-and pathway-based
analyses may help to decipher the complex genomic architecture
of MD.
Conclusions
Allelic variants rs3774937 and rs4648011 can modify hearing
outcome in patients with MD and unilateral SNHL. A patent
application number P201430716 has been submitted to the
Spanish Patent and Trademark Office.
Supporting Information
File S1 Figure S1, Scatter plot showing the principal
component analysis (PCA) in our Spanish samples compared with
different populations in HapMap. The eigenvalues for the first
three principal components accounted for most of the population
substructure in this analysis (77.5%). All individuals who were not
clustering with the main cluster (. 3 Standard deviation from
cluster center) were excluded from subsequent analysis. Using this
method we identified a total of 48 outliers individuals in our case-
control cohort. X-axis represents Principal Component 1 (PC1)
and Y-axis represents Principal Component 3 (PC3) in our
Spanish samples (diamonds), and the main populations in
HapMap: CEU, Northern European from Utah (squares),
CHB+JPB, Chinese in Beijing+ Japanese in Tokyo (triangles),
MEX (crosses), TSI, Tuscans from Italy (asterisks) and YRI,
Yoruba in Ibadan, Nigeria (circles). Table S1, Minor allelic
frequencies of the top 10 ranked signals obtained with the
Immunochip in patients with Meniere’s diasease. Table S2,
Minor allelic frequencies of 15 single nucleotide variants in the
TNFAIP3 gene in controls and patients with uni and bilateral
sensorineural hearing loss. Table S3, Minor allelic frequencies of
87 single nucleotide variants in the TNIP1 gene in controls and
patients with uni and bilateral sensorineural hearing loss. TableS4, Minor allelic frequencies of 16 single nucleotide variants in the
REL gene in controls and patients with uni and bilateral
sensorineural hearing loss. Table S5, Minor allelic frequencies
of 34 single nucleotide variants in the UBE2L3 gene in controls
and patients with uni and bilateral sensorineural hearing loss.
Table S6, Minor allelic frequencies of 9 single nucleotide variants
in the NFKB1 gene in controls and patients with uni and bilateral
sensorineural hearing loss.
(DOCX)
Acknowledgments
We would like to thank all MD patients and control subjects for
participating in this study. Sonia Cabrera is a PhD student and this work
is part of her Doctoral Thesis.
Author Contributions
Conceived and designed the experiments: SC ES NP GT PP MAR AB
JALE. Performed the experiments: SC TR. Analyzed the data: SC ES
MMB AB GT JALE. Contributed reagents/materials/analysis tools: SC
TR JB GT ASV SS EMS JF PP MAR JMES IA JALE. Wrote the paper:
SC ES TR MAR JALE.
Table 4. Effect of rs3774937 (T.C) and rs4648011 (T.G) haplotypes on hearing loss in patients with Meniere’s disease.
Time to reach .40 dB (years, mean ± SD)
HAPLOTYPE (rs3774937, rs4648011) FREQUENCY (%) UNILATERAL BILATERAL
CG 33 860.48 1060.90
GT 5 861.12 1261.20
TT 62 1060.47 1260.67
Time to reach hearing .40 dB was compared by survival curves using the Kaplan-Meier method.doi:10.1371/journal.pone.0112171.t004
NFKB1 Gene and Hearing in Meniere’s Disease
PLOS ONE | www.plosone.org 7 November 2014 | Volume 9 | Issue 11 | e112171
References
1. Clemmens C, Ruckenstein M (2012) Characteristics of Patients With Unilateral
and Bilateral Meniere’s Disease. Otol Neurotol 33(7):1266–9. doi: 10.1097/MAO.0b013e31826426b9
2. Greco A, Gallo A, Fusconi M, Marinelli C, Macri GF, et al. (2012) Meniere’sdisease might be an autoimmune condition? Autoimmun Rev 11(10):731–8. doi:
10.1016/j.autrev.2012.01.004
3. Platt M, Dilwali S, Elackattu A, Parikh JR, Stankovic KM (2013) MiningImmune Epitopes in the Inner Ear. Otolaryngology - Head and Neck Surgery
150(3):460–3. doi: 10.1177/01945998135147254. Pathak S, Goldofsky E, Vivas EX, Bonagura VR, Vambutas A (2011) IL-1b is
overexpressed and aberrantly regulated in corticosteroid nonresponders with
autoimmune inner ear disease. The Journal of Immunology 186(3):1870–9. doi:10.4049/jimmunol.1002275
5. Zhao R, Zhou H, Su SB (2013) A critical role for interleukin-1B in theprogression of autoimmune diseases. Int Immunopharmacol 17(3):658–69. doi:
10.1016/j.intimp.2013.08.0126. Svrakic M, Pathak S, Goldofsky E, Hoffman R, Chandrasekhar SS, et al. (2012)
Diagnostic and prognostic utility of measuring tumor necrosis factor in the
peripheral circulation of patients with immune-mediated sensorineural hearingloss. Arch Otolaryngol Head Neck Surg 138(11):1052–8. doi: 10.1001/
2013.jamaoto.767. Gazquez I, Moreno A, Aran I, Soto-Varela A, Santos S, et al. (2012) MICA-
STR A.4 is associated with slower hearing loss progression in patients with
Meniere’s disease. Otol Neurotol 33(2):223–9. doi: 10.1097/MAO.0b013e31824296c8
8. Requena T, Gazquez I, Moreno A, Batuecas A, Aran I, et al. (2013) Allelicvariants in TLR10 gene may influence bilateral affectation and clinical course of
Meniere’s disease. Immunogenetics 65(5):345–55. doi: 10.1007/s00251-013-0683-z
9. Gazquez I, Soto-Varela A, Aran I, Santos S, Batuecas A, et al. (2011) High
prevalence of systemic autoimmune diseases in patients with Meniere’s disease.PLoS One 6(10):e26759. doi: 10.1371/journal.pone.0026759
10. Tyrell J, Whinney DJD, Ukoumunne OC, Fleming LE, Osborne NJ (2014)Prevalence, associated factors and comorbid conditions for Meniere’s disease.
Ear Hearing 35(4):e162–9. doi: 10.1097/AUD.0000000000000041
11. Caamano J, Hunter CA (2002) NF-kappaB family of transcription factors:central regulators of innate and adaptative immune functions. Clin Microbiol
Rev 15(3): 414–429. doi: 10.1128/CMR.15.3.414-42912. Huang TT, Kudo N, Yoshida M, Miyamoto S (2000) A nuclear export signal in
the N-terminal regulatory domain of IkappaBalpha controls cytoplasmiclocalization of inactive NF-kappaB/IkappaBalpha complexes. Proc Natl Acad
Sci USA 97 (3) 1014–1019. doi:10.1073/pnas.97.3.101
13. Hoesel B, Schmid JA (2013) The complexity of NF-kN signaling in inflammationand cancer. Mol Cancer 12: 86 doi:10.1186/1476-4598-12-86.
14. Parkes M, Cortes A, van Heel DA, Brown MA (2013) Genetic insights intocommon pathways and complex relationships among immune-mediated
diseases. Nat Rev Genet 14(9):661–73. doi:10.1038/nrg3502
15. Committee on Hearing and Equilibrium. Guidelines for the diagnosis andevaluation of therapy in Meniere’s disease (1995) American Academy of
Otolaryngology-Head and Neck Foundation, Inc. Otolaryngol Head Neck Surg113: 181–185.
16. Purcell S, Neale B, Todd-Brown K, Thomas L, Ferreira MA, et al. (2007)PLINK: a tool set for whole-genome association and population-based linkage
analyses. Am J Hum Genet 81:559–75. doi.org/10.1086/519795
17. Ward LD, Kellis M (2012) HaploReg: a resource for exploring chromatin states,conservation, and regulatory motif alterations within sets of genetically linked
variants. Nucleic Acids Res 40:D930–4. doi: 10.1093/nar/gkr91718. Xia K, Shabalin AA, Huang S, Madar V, Zhou YH, et al. (2012) seeQTL: a
searchable database for human eQTLs. Bioinformatics 28(3):451–2. doi:
10.1093/bioinformatics/btr67819. Boyle AP, Hong EL, Hariharan M, Cheng Y, Schaub MA, et al. (2012)
Annotation of functional variation in personal genomes using RegulomeDB.Genome Research 22(9):1790–1797. doi: 10.1101/gr.137323.112
20. Campbell CA, Santina Della CC, Meyer NC, Smith NB, Myrie OA, et al. (2009)
Polymorphisms in KCNE1 or KCNE3 are not associated with Meniere diseasein the Caucasian population. Am J Med Genet A 152A(1):67–74. doi: 10.1002/
ajmg.a.3311421. Hietikko E, Kotimaki J, Okuloff A, Sorri M, Mannikko M (2012) A replication
study on proposed candidate genes in Meniere’s disease, and a review of thecurrent status of genetic studies. Int J Audiol 51(11):841–5. doi: 10.3109/
14992027.2012.705900
22. Gazquez I, Moreno A, Requena T, Ohmen J, Santos-Perez S, et al. (2012)Functional variants of MIF, INFG and TFNA genes are not associated with
disease susceptibility or hearing loss progression in patients with Meniere’sdisease. Eur Arch Otorhinolaryngol 270(4):1521–9. doi: 10.1007/s00405-012-
2268-0
23. Lander ES, Linton LM, Birren B, Nusbaum C, Zody MC, et al (2001) Initial
sequencing and analysis of the human genome. Nature 409:860–921.doi:10.1038/35057062
24. Int. HapMap Consort (2005) A haplotype map of the human genome. Nature437:1299–1320. doi:10.1038/nature04226
25. Ricano-Ponce I, Wijmenga C (2013) Mapping of Immune-Mediated Disease
Genes. Annu Rev Genom Human Genet 14(1):325–53. doi: 10.1146/annurev-genom-091212-153450
26. Trynka G, Hunt KA, Bockett NA, Romanos J, Mistry V, et al. (2011) Densegenotyping identifies and localizes multiple common and rare variant association
signals in celiac disease. Nature Genetics 43(12):1193–201. doi: 10.1038/ng.998
27. Cooper JD, Simmonds MJ, Walker NM, Burren O, Brand OJ, et al. (2012)
Seven newly identified loci for autoimmune thyroid disease. Hum Mol Genet
21(23):5202–8. doi: 10.1093/hmg/dds357
28. International Multiple Sclerosis Genetics Consortium (IMSGC), Beecham AH,
Patsopoulos NA, Xifara DK, Davis MF, et al. (2013) Analysis of immune-relatedloci identifies 48 new susceptibility variants for multiple sclerosis. Nat Genet
45(11):1353–60. doi:10.1038/ng.2770
29. Ellinghaus D, Baurecht H, Esparza-Gordillo J, Rodrıguez E, Matanovic A, et al.
(2013) High-density genotyping study identifies four new susceptibility loci for
atopic dermatitis. Nat Genet 45(7):808–12. doi: 10.1038/ng.2642
30. Hughes T, Coit P, Adler A, Yilmaz V, Aksu K, et al. (2013) Identification of
multiple independent susceptibility loci in the HLA region in Behcet’s disease.Nat Genet 45(3):319–24. doi: 10.1038/ng.2551
31. Lopez-Escamez JA, Vilchez JR, Soto-Varela A, Santos-Perez S, Perez-Garrigues
H, et al. (2007) HLA-DRB1*1101 allele may be associated with bilateralMeniere’s disease in southern European population. Otol Neurotol 28(7):891–5.
doi: 10.1097/MAO.0b013e3180dca1cc
32. Lopez-Escamez JA, Saenz-Lopez P, Acosta L, Moreno A, Gazquez I, et al.
(2010) Association of a functional polymorphism of PTPN22 encoding alymphoid protein phosphatase in bilateral Meniere’s disease. Laryngoscope
120:103–107. doi: 10.1002/lary.20650
33. Zhu J, Mohan C (2010) Toll-Like Receptor Signaling Pathways—TherapeuticOpportunities. Mediators of Inflammation 2010:781235. doi: 10.1155/2010/
781235
34. Waldner H (2009) The role of innate immune responses in autoimmune disease
development. Autoimmun Rev 8(5): 400–4. doi: 10.1016/j.autrev.2008.12.019
35. Akira S, Takeda K (2004) Toll-like receptor signalling. Nat Rev Immunol 4(7):
499–511. doi:10.1038/nri1391
36. Adams JC, Seed B, Lu N, Landry A, Xavier RJ (2009) Selective activation ofnuclear factor kappa B in the cochlea by sensory and inflammatory stress.
Neuroscience 160(2): 530–9. doi: 10.1016/j.neuroscience.2009.02.073
37. Hunt KA, Mistry V, Bockett NA, Ahmad T, Ban M, et al. (2013) Negligible
impact of rare autoimmune-locus coding-region variants on missing heritability.
Nature 498(7453): 232–5. doi:10.1038/nature12170
38. Tryka KA, Hao L, Sturcke A, Jin Y, Wang ZY, et al. (2014) NCBI’s Database of
Genotypes and Phenotypes: dbGaP. Nucleic Acids Res 42:D975–9. doi:10.1093/nar/gkt1211
39. Nair RP, Duffin KC, Helms C, Ding J, Stuart PE, et al. (2009) Genome-widescan reveals association of psoriasis with IL-23 and NF-kappaB pathways. Nat
Genet 41(2):199–204. doi: 10.1038/ng.311
40. Li X, Ampleford EJ, Howard TD, Moore WC, Torgerson DG, et al. (2012)Genome-wide association studies of asthma indicate opposite immunopatho-
genesis direction from autoimmune diseases. J Allergy Clin Immunol130(4):861–8. doi: 10.1016/j.jaci.2012.04.041
41. Bowes J, Orozco G, Flynn E, Ho P, Brier R, et al. (2011) Confirmation ofTNIP1 and IL23A as susceptibility loci for psoriatic arthritis. Ann Rheum Dis
70(9):164. doi: 10.1136/ard.2011.150102.1
42. Stahl EA, Raychaudhuri S, Remmers EF, Xie G, Eyre S, et al. (2010) Genome-wide association study meta-analysis identifies seven new rheumatoid arthritis
risk loci. Nat Genet 42(6):508–14. doi: 10.1038/ng.582
43. Chung SA, Taylor KE, Graham RR, Nititham J, Lee AT, et al. (2011)
Differential genetic associations for systemic lupus erythematosus based on anti-
dsDNA autoantibody production. PLoS Genet 7(3):e1001323. doi: 10.1371/journal.pgen.1001323
44. Han JW, Zheng HF, Cui Y, Sun LD, Ye DQ, et al. (2009) Genome-wideassociation study in a Chinese Han population identifies nine new susceptibility
loci for systemic lupus erythematosus. Nat Genet 41(11):1234–7. doi:10.1038/ng.472
45. Teslovich TM, Musunuru K, Smith AV, Edmondson AC, Stylianou IM, et al.
(2010) Biological, clinical and population relevance of 95 loci for blood lipids.Nature 466(7307):707–13. doi: 10.1038/nature09270
46. Liu JZ, Almarri MA, Gaffney DJ, Mells GF, Jostins L, et al. (2012) Dense fine-mapping study identifies new susceptibility loci for primary biliary cirrhosis.
Nature Genet 44(10):1137–41. doi:10.1038/ng.2395
NFKB1 Gene and Hearing in Meniere’s Disease
PLOS ONE | www.plosone.org 8 November 2014 | Volume 9 | Issue 11 | e112171